Wound healing and cutaneous scarring models of the human skin

Wound healing and cutaneous scarring models of the human skin

9

Mohammed Ashrafi*,†,‡,#, Adam Hague*,#, Mohamed Baguneid†, Teresa Alonso-Rasgado‡, Ardeshir Bayat*,†,‡ * Plastic & Reconstructive Surgery Research, Centre for Dermatological Research, University of Manchester, Manchester, United Kingdom, †University Hospital of South Manchester NHS Foundation Trust, Manchester, United Kingdom, ‡Bioengineering Group, School of Materials, University of Manchester, Manchester, United Kingdom

1 Introduction Wound healing is a multifactorial process, which commences after any form of cutaneous injury resulting in immediate vasoconstriction and coagulation [1]. The subsequent release of cytokines by keratinocytes and platelets leads to initial inflammation, with various immune cells, including neutrophils and monocytes, migrating into the wound bed [1]. Release of growth factors by macrophages stimulates angiogenesis, resident epithelial cells to proliferate, and fibroblast infiltration to occur, ultimately assisting with wound closure [2]. The final visual appearance of the wound-healing process is of upmost importance to both patients and physicians alike, with adverse scarring being a cause for concern. A major challenge currently facing researchers investigating this area is creating an ideal model to study this complex process, evaluate candidate therapies, and ultimately improve aesthetic and functional outcomes for patients with abnormal wounds and scars [3]. Many models currently exist, investigating all stages of wound healing including reepithelialization and scar remodeling [4]. Animal models have been widely used in attempts to replicate the human skin [5]. Pig skin is currently thought to be the most accurate animal model; however, no animal skin is an exact replica of the human skin, leading to significant differences in the wound-healing mechanisms between them [6,7]. Additionally, ethical issues surrounding their use also need to be taken into consideration [8]. Numerous human scar and wound-healing models have been developed, including in silico, in vitro, ex vivo, and in vivo. The aim of this review is to summarize and evaluate each of the above models including their advantages and limitations (Table 1).

2 In silico The large variety of interactions that influence wound healing requires the need for computational models in order to accurately understand and investigate this complex process #

Authors contributed equally to the chapter and are joint first authors.

Skin Tissue Models. https://doi.org/10.1016/B978-0-12-810545-0.00009-7 © 2018 Elsevier Inc. All rights reserved.

Human wound models

Model

Description/features

Application

Advantages

Disadvantages

- The use of mathematical equations based on known cellular behaviors to simulate wound healing - A computational modeling technique that can simulate interactions between cells and their environment. Classically consist of three components: agent, region, and patch

- Investigation of all stages of wound healing, along with the impact that various cells, cytokines, and growth factors have on this process

- Can be used as an initial screening tool to help plan clinical trials/ research

- Can involve complex mathematical equations - The lack of standardized outcome measures - Remain experimental unless validated

- A defect simulating a wound is created in cells cultured in monolayer - Coculture of more than one cell type. The use of the Transwell system has classically been described for this purpose - Cells cultured on a threedimensional (3-D) scaffold or grown in a 3-D format in order to create a skin “equivalent”

- Studying rates of cell migration and proliferation in a wound setting, for example, keratinocytes and fibroblasts - Investigation of specific cellular interactions and subsequent effects - Investigation of scar pathogenesis

- Single-cell models are relatively quick and inexpensive to conduct and allow the investigation of one specific cell type - Ability to include more than one cell type more accurately reflects the wound environment, particularly with organotypic models

- Models cannot accurately reflect cellular interactions in the wound environment, particularly single-cell and coculture models - Many models use serum, which is inflammatory when compared with plasma and interstitial fluid. Can therefore be difficult to translate findings into the in vivo setting - The lack of validated biomarkers - Dependence on excised tissue

202

Table 1 

In silico Differential equations Agent-based models

In vitro Single cell Coculture Organ culture

Skin Tissue Models

Scars Skin with the epidermis intact The skin with artificially created wounds

- The use of harvested cutaneous tissue - Common scar tissue used includes that from keloids and striae albae - Wounds created can be full thickness or partial thickness and include burns

- Investigation of scar pathogenesis and physiology of wound healing - Testing of potential therapeutic agents - Genotoxicity testing

- Provides a safe environment for the assessment of potential treatments - Creates an environment that more closely resembles normal skin

- Short investigative opportunity - Not subject to the same environmental factors that would be present in a patient - The lack of standardization in study design

- The use of live subjects. These can be either patients with wounds or scars already present or volunteers in which a wound is created

- Validation of wound and scar therapies

- Physiology of healing is identical to that in patients

- The lack of standardization in study design - Difficulties with recruitment

In vivo Scars Wounds

Wound healing and cutaneous scarring models of the human skin203

ex vivo

204

Skin Tissue Models

Fig. 1  Diagram illustrating examples of in silico human models of scar formation and wound healing.

(Fig. 1) [9]. Numerous work has been conducted into designing such models for various stages of the wound-healing processes [9]. These can then subsequently be validated experimentally and used to assist in various areas of tissue engineering research. Differential equations have been widely used to investigate the differing phases of wound healing [9]. Models specifically looking at the inflammatory phase of healing have used ordinary differential equations to outline the effects of variability between patients in growth factors and structural proteins, along with macrophages and fibroblasts [10,11]. Equations have also been used to model wound closure [12,13]. Arciero et al. investigated the mechanisms that coordinate cell migration during wound healing by deriving a two-dimensional continuum model [12]. Evolution equations were then solved numerically using a level set method. Mathematical models were also utilized by Wearing et al. who looked into the effects of keratinocyte growth factor on reepithelialization [13]. The final stage of the wound-healing process, tissue remodeling, was investigated by McDougall et al. [14]. They applied mathematical models to demonstrate the influences that constituents of the extracellular matrix (ECM) have on the trajectories of fibroblasts and direction of collagen deposition. Agent-based modeling is a computational modeling technique that can simulate complex behaviors and interactions between agents and their environment while taking into consideration the stochastic nature of biological processes [15]. Agent-based models (ABMs) typically consist of three components: agent, region, and patch [15]. The agent represents cells or molecules that move in the region, which consists of individual patches. The simulated behaviors of agents in their virtual environment are based on current knowledge regarding their known behavior [16]. They have several advantages including being easier to use for nonmathematicians while also simplifying the representation of tissue structure [16]. Adra et al. derived an ABM to assess the regulatory actions transforming growth factor-beta 1 (TGF-β1) has in tissue regeneration [17]. They subsequently validated the model by using it to create a virtual piece of epidermis and comparing keratinocyte behavior and the actions of TGF-β1 with the existing literature. Similarly, Li et al. used a three-dimensional (3-D) ABM of the epidermis to investigate epithelial renewal and basal keratinocyte regeneration [18].

Wound healing and cutaneous scarring models of the human skin205

ABMs have also been used to provide a framework for the creation of multiscale models for wound healing. Along with investigating the effects of interactions between keratinocytes and fibroblasts [19], such models have also been used to study the biological influences that regulate human keratinocyte organization [20]. Anderson et al. advanced this type of model to create a hybrid continuous-discrete model for deterministic systems [21], which has been utilized to study fluid flow through vascular networks [22,23].

3 In vitro In vitro models have been widely used in the investigation of wound healing and scar pathogenesis [3]. They are generally inexpensive and require less ethical considerations when compared with their in vivo counterparts [24]. They vary in complexity ranging from single-cell and coculture (two different cell types) models to organotypic multicellular-layered constructs (Fig. 2) [24]. Single-cell models commonly involve culturing cells in monolayer directly on the surface of the culture dish itself or on substrates including collagen and fibrin [24]. A defect, simulating a wound, is then created in the cell monolayer through mechanical [25,26] or chemical methods [27]. The defect is then gradually repopulated by the surrounding cells [28]. Using microscopy, the rates at which this proliferation and migration occur can then be determined. Such models classically involve keratinocytes [29] or fibroblasts [28,30]. Although they are relatively quick and inexpensive to conduct, single-cell models do not accurately reflect the wound environment [3]. In order to overcome this, several other models have been created. Cocultures of fibroblasts and keratinocytes using Transwell systems (Corning Costar, MA) have been conducted, allowing the

Fig. 2  Diagram illustrating examples of in vitro human models of scar formation and wound healing.

206

Skin Tissue Models

investigation of the interactions between these two cell types and the subsequent effects [31,32]. The Transwell assay is composed of two separate mediums divided by a porous membrane and can be used for various cell types including epithelial [33] and mesenchymal cells [34]. In vitro 3-D models have also been used in attempts to construct and mimic the ECM while also more accurately reflecting normal wound physiology [35]. It has already been shown that both collagen and collagenase production by fibroblasts varies when they are cultured in two and 3-D, presumably due to the action of integrins [36]. Nayak et al. utilized a construct of sericin to assess the proliferation and viability of cocultures of keratinocytes and fibroblasts, which were located on the upper and lower surfaces, respectively [37]. Similarly, collagen and fibrin have been used to create a 3-D environment, with both promoting fibroblasts to differentiate toward the scar phenotype when combined with a mechanical load [38]. Additionally, Karamichos et al. created a 3-D collagen matrix to study the effects of different culture conditions on matrix remodeling and fibroblast migration [39]. The knowledge that extensive interaction between keratinocytes and fibroblasts regulates the production of dermal ECM, led to the development and use of organotypic skin equivalents in scar research [40]. Bellemare et al. isolated keratinocytes from hypertrophic scars on a dermal matrix and created a fully differentiated epidermis [41]. This subsequently exhibited characteristics of adverse scarring and highlighted the contribution that keratinocytes have in hypertrophic scar formation. Likewise, Chiu et al. investigated the effects of photodynamic therapy on keloid scars through the use of an organotypic skin equivalent with keratinocytes and keloid fibroblasts [42]. The use of such models to investigate potential therapeutics is, however, limited due to the need for excised scar tissue along with a lack of validated biomarkers [3]. This led to the development of a hypertrophic scar model created from adipose-derived mesenchymal stem cells [43]. This model exhibited many similar characteristics to that of hypertrophic scars (e.g., increased collagen I secretion) while also allowing for the identification of measurable scar parameters that were subsequently validated using various treatments. The authors did, however, conclude that this model is only relevant for hypertrophic scars caused from injury where the adipose tissue is exposed.

4 Ex vivo Ex  vivo models initially involve the harvesting of cutaneous tissue from informed volunteers and with appropriate research ethical approvals in place. These models are critical in evaluating the pathophysiology of scars and wounds and their response to treatment, especially in diseases, such as keloid, where no animal models exist. The three main categories of cutaneous ex vivo models published are scars [44–47], skin with the epidermis intact [48–53], and skin with various artificially created wounds [48,52,54–58] (Fig. 3). Scar models include keloid [44–47], striae alba [46], hypertrophic [46], and fineline scars [46], with the site of scar harvest differing greatly between studies [46,47]. Biopsies of the scars used varied in shape and size; however, the majority were ­created

Wound healing and cutaneous scarring models of the human skin207

Fig. 3  Diagram illustrating examples of ex vivo human models of scar formation and wound healing.

using circular punch biopsies [44,46,47], although square-shaped explants have also been utilized [45]. The minimum diameter of biopsies used was 3 mm [44] with the largest reported at 6 mm [44,46]. Both Hanks buffer [44,45,47] and phosphate-­ buffered saline [46] have been used to wash the explants prior to embedding. Using multiple validation measures including histological assessment, cell proliferation and apoptotic assays, ELISA, and immunohistochemical and immunofluorescence staining, both Duong et al. [45] and Bagabir et al. [44] were able to preserve keloid ex vivo organ culture models for up to 6 weeks. Various media were used to support the explants (Table 2), with the use of supplemented William’s E medium changed every 3 days, supporting superior proliferation with reduced cytotoxicity [44]. All explant models were embedded in collagen gel matrix with the epidermis either submerged in media [44,45] or air-exposed [44–47], with superior morphology being observed in the air-exposed explants [45]. Bagabir et  al. [44] supplemented the media with dexamethasone, which is employed in the clinical management of keloid disease, and found shrinkage compared with untreated controls at 4  weeks, whereas at 3 weeks, no such macroscopic changes were apparent, highlighting the need for keloid models with longevity in order to accurately determine treatment output. The same model has also been used to evaluate photodynamic [46] and antifibrotic therapies [47]. Skin explant models with the epidermis intact are employed as a control to wounded explant models. They are typically harvested from the breast [49–52] and abdomen of patients undergoing elective surgery [51,53]. The size of the explants described

Table 2 

Ex vivo models

Model

Shape/ size

Embedding techniques

Epidermis

Culture duration

Culture media

Validation techniques -LDH -MTT -TUNEL -Ki67 -Col I–III ELISA -Histology -IHC-IF -IHC -TUNEL -ISH

-Histology -IHC -IF -qRT -PCR -LDH -MTT -TUNEL -Ki67 -Col I–III ELISA -Histology -IHC -qRT -PCR

[44]

-Scar

-Circle -3–6 mm

-RTCG

-AE-S

-6 weeks

-DMEM -WEM -Changed every 3 days

[45]

-Scar

-Square -3–5 mm

-RTCG

-AE -S

-6 weeks

[46]

-Scar

-Circle -6 mm

-RTCG

-AE

-7 days

-DMEM -FBS -Changed every 24 -72 h -WEM

[47]

-Scar

-Circle -4 mm

-RTCG

-AE

-4 weeks

-WEM -Changed every 3 days

Therapeutics investigated

Disinfection

-Sternum -Arm -Pubis -Scalp -Face -Ear -Abdomen

-DEX

-HB

-NS

-Nil

-HB

-Abdomen -Arm -Breast -Thorax -Ear -Sternum -Arm -Pubis -Scalp -Face -Ear -Abdomen

-PDT

-PBS

-EGCG -PAI-1 -DEX

-HB

Site

Model

Shape/ size

Embedding techniques

Epidermis

Culture duration

Culture media

Validation techniques

-Circle -6 mm -Circle -6 mm -3  mm FTW -Circle -3 mm -Square -1.5 cm2

-RTCG -RTCG

-AE -AE

-3 weeks -3 weeks

-DMEM -Changed every 48 h

-TF

-AE

-24 h

-TF

-AE

-15 days

[48]

-Skin -Wound

[49]

-Skin

[50]

-Skin

[51]

-Skin

-Circle -16 mm

-NI

-AE

-24 h

[52]

-Skin -Wound

-NS

-AE

-6 days

[53]

-Skin

-Square -1 cm2 -Square -1 cm2 -Circle -4  mm FTW -NS

-CS

-AE

-14 days

[54]

-Wound

-TF

-AE

-21 days

-Circle -8 mm -4  mm FTW

Site

Therapeutics investigated

-LDH -Histology -IHC -VEGF ELISA

-NS

-IL 10

-HB -70% ethanol

-IMDM

-IHC

-Breast

-NS

-EpiLife with EDGS -Changed every 36 h -DMEM -Changed every 24 h -KBM

-Histology -TUNEL

-Breast

-Cytokines -LPS -Fixator pin -Biomaterial

-Comet assay -MTT -Planimetry -Histology

-Breast -Abdomen

-Toxins

-HB

-Breast

-Toxins

-NS

-Abdomen

-Nil

-Betadine -70% ethanol

-Abdomen

-Hyaluronan

-NS

-DMEM -Changed every 48 h -DMEM -Changed NS

-Histology -TUNEL -Ki-67 -IF -Histology

Disinfection

-PBS

(Continued)

Table 2 

Continued

Model [55]

-Wound

[56]

-Wound

[57]

-Wound

[58]

-Wound

Shape/ size

Embedding techniques

Epidermis

Culture duration

Culture media

Validation techniques

-Circle -6 mm -3  mm FTW -Square -Burn wound -Circle -6 mm -3  mm PTW -Circle -8 mm -4  mm FTW

-Nil

-S

-14 days

-DMEM -Changed every 48 h

-Culture disc

-AE

-7 days

-RTCG

-AE

-16 days

-DMEM -Changed every 48 h -DMEM -Changed every 48 h

-Circle -8 mm -2  mm FTW

-TF

-S and AE

-8 days

-NS -Changed NS

Site

Therapeutics investigated

Disinfection

-Histology -Cell culture

-Breast

-Nil

-NS

-IHC -ISH

-Trunk

-Keratinocytes

-NS

-Histology -IHC -qRT-PCR -Western blotting -LDH -Flow cytometry -Histology -IHC

-Abdomen

-Electric stimulation

-PBS

-Tissueengineered human skin

-Dermal substitutes

-NS

FTW, full-thickness wound; NS, not specified; PTW, partial-thickness wound; RTCG, rat tail collagen gel; TF, Transwell filter; NI, Netwell inserts; CS, cell strainer; AE, air-exposed; S, submerged; DMEM, Dulbecco’s modified eagle’s medium; WEM, William’s E medium; FBS, fetal bovine serum; IMDM, Iscove’s modified Dulbecco’s medium; EDGS, EpiLife defined growth supplement; KBM, keratinocyte basal medium; LDH, lactate dehydrogenase; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling; Col, collagen; ELISA, enzyme-linked immunosorbent assay; IHC, immunohistochemistry; IF, immunofluorescence; ISH, in situ hybridization; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; VEGF, vascular endothelial growth factor; DEX, dexamethasone; PDT, photodynamic therapy; EGCG, (−)-epigallocatechin-3-gallate; PAI-1, plasminogen activator inhibitor-1; IL, interleukin; LPS, lipopolysaccharides; HB, Hanks buffer; PBS, ­phosphate-buffered saline.

Wound healing and cutaneous scarring models of the human skin211

varies in size and shape, ranging from 6 to 16 mm circular punches [48,49,51], uniform squares with a surface area of 1–1.5 cm2 [52] or being irregularly shaped [50]. Numerous solutions have been used to disinfect the explants including 70% ethanol [48,53], betadine [53], Hanks buffer with [51] or without antibiotic supplementation [48], and phosphate-buffered saline supplemented with gentamicin and amphotericin B [50]. Explant viability ranged from 24 h [49] to 3 weeks [48] with the epidermis being air-exposed in all models [48–53]. Balaji et al. [48] embedded the skin explants in rat tail collagen 1 gel matrix cultured in supplemented serum-free Dulbecco’s modified eagle’s medium with the epidermal surface air-exposed. With the media changed every other day for 21 days, they showed detachment and decreasing thickness of the epidermis at day 14; however, they showed the dermis was similar to day 0 skin and lactate dehydrogenase (LDH) activity declined from day 1 to day 21. Companjen et al. used a different embedding technique with the skin explants inserted into Transwell filters, with the dermis suspended in culture medium and the epidermis also air-exposed [49]. These biopsies, taken from healthy volunteers and those with psoriatic plaque-type skin lesions, were incubated for only 24 h in standard conditions (5% carbon dioxide and 37°C) or special conditions, where explants were placed into Tedlar culture bags with an artificial atmosphere of 5% carbon dioxide and 95% oxygen at 32°C. They found morphological deterioration of psoriatic lesions in standard conditions that were abolished under special conditions. No such deterioration of normal skin explants was noted. They concluded the model allowed maintenance of the normal skin architecture without spontaneous induction of regenerative maturation markers, through measurement of cytokine release with ELISA and immunohistochemistry. However, they only cultured the model for a very short time period [49]. Xu et al. anchored a partial dermal thickness skin explant using sutures to a nylon mesh cell strainer and showed normal epidermal morphology, with keratinocyte proliferation up to 12 days and an intact basement membrane for up to 21 days. [53]. The epidermis was air-exposed, while the dermis was surrounded in media that was changed every other day. The skin explant models have been used to assess cytokine activity [48,49], glycosaminoglycans [50], and toxins [51,52]. Cutaneous explants with iatrogenic wounds were either harvested from the breast [52,55], the abdomen following abdominoplasty surgery [54,57], or the trunk [56]. All but one study used circular skin explants 6–8 mm in diameter [48,54–58]. TomicCanic et al. used 1 cm2 explants with a 4 mm circular punch biopsy through the reticular dermis [52]. All other studies have also utilized circular wounds within the explants ranging in depth from 1 mm [55], the upper dermis [56], to full thickness [48,54,57,58]. Kratz et al. also created a deep dermal burn wound model by applying a heated brass string to the epidermal surface for 1 s at a temperature of 150°C [55]. Disinfection, embedding, length of culture, type of media, and validation methods varied between the studies (Table  2). Balaji et  al. evaluated cytokine treatment on wound healing by embedding the wounded explants in rat tail collagen matrix, maintaining the model for up to 3 weeks with media changed every other day and monitoring viability using a LDH cytotoxicity assay and histological evaluation [48]. Kratz et al. found discrepancies in the wound-healing process between incisional and burn wounds, highlighting the vital role ex vivo models have in understanding the normal

212

Skin Tissue Models

healing process of different types of wounds, and thus potentially allowing the investigation of therapeutics that are wound type-specific [55]. These cutaneous wounded explant models have also been used to assess toxins [52], glycosaminoglycans [54], and keratinocyte transplantation [56]. Dermal substitutes have also been assessed by inserting them into the full-thickness wounds created in the tissue explants [54,58]. Sebastian et al. used the donut-shaped full-thickness wounded explants to investigate the effect of electric stimulation on wound closure and found enhanced epidermal proliferation [57]. The use of these diverse scar and wound models described above in assessing cytokine activity, toxin effects, cell migration, and proliferation is extensively highlighted. These models have also allowed assessment of other features such as epidermal thickness [44], epithelialization [57], and the role of immune components in healing and response to treatment [5]. Other features of wound healing where these models could be used to investigate include the study of dermal thickness, angiogenesis, hair follicles, rete ridge formation, and ECM proteins.

5 In vivo In vivo cutaneous scar and wound models are ideal for investigating the pathophysiology of healing in live subjects (Fig. 4). In vivo models are used to either assess normal scarring or wound-healing processes or used to investigate the effect of candidate therapeutics [59,60]. To assess the normal scarring process, the creation of a scar in healthy volunteers is a model that has been utilized. Dunkin et al. hypothesized that there is a critical depth of wounding beyond which a fibrous scar develops [61]. They confirmed this by creating an incisional wound of varying depth with deep dermal at one end and superficial dermal at the other in 113 healthy volunteers along the lateral aspect of the hip. Using digital images and high-frequency ultrasound, they found no detectable scar at the shallow end of the wound at the study end point of 9 months, confirming their hypothesis [61]. The impact of treatments on scars has been assessed in clinical studies. Ud-din et al. conducted a case series of 65 patients with keloid disease to assess steroid-sensitive responders using mostly subjective parameters [59]. Over a 3-month period, they found keloid scar steroid sensitivity was associated with anatomical location and frequency of injections. However, the lack of objective measures of response and short follow-up period are limitations to be considered. The same group conducted a similar case series on patients with keloid disease undergoing photodynamic therapy over a longer followup period of 9 months; however, on this occasion, they also included quantifiable noninvasive objective measures of response including spectrophotometric intracutaneous analysis and full-field laser perfusion imagery [60]. Human in  vivo scars have also been used to assess less conventional treatments such as electric stimulation [62]. The application of electric stimulation on raised dermal scars found reduced pain and pruritus over a 2-month study period in 19 patients [62]. Although, objective measurements of scar melanin, collagen, hemoglobin, and blood flow were measured, the

Wound healing and cutaneous scarring models of the human skin213

Fig. 4  Diagram illustrating examples of in vivo human models of scar formation and wound healing.

p­ rimary ­significant findings were subjective outcomes. As well as noninvasive objective ­measures, invasive biopsies of scars following treatment have also been done to assess morphological and immunohistochemical changes post treatment [63]. Various cutaneous wound models in patients and healthy volunteers have been investigated with regards to the healing process and the effect of various treatment modalities. These models include superficial [64–69], burn [70,71], incisional [72], and excisional wounds [73–77]. Superficial wounds include the blister model [64–66], tape-stripping method [67], and skin graft donor sites [68,69]. The blister model allows the creation of a standardized superficial wound that can be monitored for epidermal regeneration. This is created using various suction devices, and these models can be used to assess drug effects and drug absorption [65,66]. They are easily replicated; however, they are limited to only providing a useful tool for the investigation of superficial rather than deep dermal wounds. Less detrimental to the epidermis is the tape-stripping method. This involves the noninvasive application of adhesive tape to the skin with repeated stripping of the epidermis that leads to skin barrier loss [67]. Although offering the least side effects and proving useful in assessing treatment penetration, these advantages are limited to

214

Skin Tissue Models

the stratum corneum. Skin graft donor sites are a further model of superficial wounding [68,69]. These are advantageous as they are easily made and duplicate wounds can be created in the same individual that allows the investigation of treatments with optimum control comparisons. They do, however, heal rapidly making a significant detectable difference difficult. The creation of a burn wound in healthy volunteers is another option. Mattsson et al. produced a superficial burn injury on the flexor surface of the forearm using an electrically heated aluminum rod at 51°C and assessed the impact of intravenous lidocaine infusion on the inflammatory phase of healing [71]. They found a significantly faster restitution of residual erythema in the active group compared with placebo at 12 h post burn. Another method of creating a burn wound is the use of ultraviolet radiation [70]. The application of this appears a safe model, with the only long-term consequence being tanning at the irradiated site. With the application of three times the minimal erythema dose, Bishop et al. found no blistering and no loss of skin barrier function [70]. Other in vivo models of wound healing include incisional and excisional wounds. Incisional cutaneous wound models involve the immediate primary closure of the wound following its creation and can be used to assess wound healing and comparison of standard versus investigative treatments. Conde-Green et  al. assessed the use of negative-pressure wound therapy in patients who underwent primary abdominal wall closure and found a significant reduction in wound complications and dehiscence rates compared with conventional dry gauze dressings [72]. The use of a biopsy punch of various sizes to create a full-thickness excisional wound in healthy individuals allows, with the use of invasive and noninvasive techniques, the profiling of normal healing and the impact of treatments. Biopsies are typically taken at multiple time points usually from the inner aspect of the upper arms under local anesthetic [73,77]. This makes the process convenient for the subjects and the location being relatively well hidden. Ud-din et al. used objective noninvasive devices such as spectrophotometric intracutaneous analysis, full-field laser perfusion, and 3-D imaging to ascertain melanin, hemoglobin, collagen, blood flow, and wound size up to 14 days post wounding [74]. They found corroboration of these findings with immunohistochemical techniques, thus supporting the use of these noninvasive devices in wound theranostics. The excisional punch biopsy model has also been used to assess treatments such as skin substitutes [75,76] and also less common treatments such as electric stimulation [73,77]. Greaves et al. assessed the use of skin substitutes including Decellularised Dermis (NHSBT, Watford, the United Kingdom), Integra Matrix Wound Dressing (Integra LifeSciences Inc., Plainsboro, NJ), and autografting of the skin and found enhanced angiogenesis [75] with acute wounds treated with Decellularised Dermis and reduced dermal fibrosis compared with control wounds [76]. Ud-din et al. identified increased blood flow in this excisional wound model following electric stimulation of the wound using noninvasive objective devices [77] and later confirmed this using immunohistochemistry and western blotting [73]. They also found increased angiogenic marker response in electrically stimulated wounds [73]. In  vivo models using animals have also been widely described in the investigation of wound healing. Perhaps the most economical of these is the murine model.

Wound healing and cutaneous scarring models of the human skin215

Numerous studies have used mice, with wounds being created on the head, ear, tail, and back [78,79]. Mice also have the additional benefit of being amenable to genetic manipulation [80]. These genetically modified mice allow for specific genes and their function to be investigated during the wound-healing process [80]. Despite their benefits, wound healing in mice primarily occurs through contraction and is therefore significantly different from that in humans [79,80]. Attempts have been made to try and overcome this limitation, however, through the use of back splints to limit the degree of wound contraction [81]. The animal model that currently offers the most accurate replication of human wound healing is that of the pig [82]. They have a thick dermis, which is structurally similar to that in humans [83], and have sparse hair and skin that is adherent to underlying structures, which is also observed in humans [84]. The porcine model has been used to study various wound-healing pathologies including chronic ischemic wounds [84] and burns [85]. They have also been instrumental in the development of epidermal and dermal skin substitutes during preclinical studies [86,87]. However, despite their benefits, there are several limitations to their use. Despite being similar to the human skin, there are significant structural differences observed in porcine skin, such as a less vascular dermis [88]. Pigs are also more expensive and difficult to manage when compared with their murine counterparts. Additionally, as with all animal models, there are numerous ethical considerations that must also be taken into account before their use.

6 Conclusion Human cutaneous scar and wound models are vital in helping researchers better understand the physiological and pathological processes of normal and abnormal ­healing, and in certain scenarios can be considered to be superior to animal models in the absence of an exact replica. In silico, in vitro, ex vivo, and in vivo models also provide relevant and realistic models to assess the therapeutic effects of experimental treatments aimed at improving the wound-healing processes. The use of in silico models are currently limited by the absence of standardization and the lack of validation and therefore remain in the experimental phase [89]. The need for transferability of methods across different laboratories and inter- and intralaboratory reproducibility are crucial before such models are deemed reliable [8]. In vitro models are beneficial as they are relatively less expensive and their complexity is broad ranging from single-cell models to organ cultures. However, translation of findings into an in vivo model is challenging given the multiple confounding factors encountered in clinical situations. To some extent, ex  vivo human models aim to solve these difficulties in clinical translatability. They provide an excellent means of assessing the complex wound-healing processes and providing a platform to assess the impact of treatments in a safe environment. However, certain environmental factors patients with wounds would be subjected to are not reproducible, and these models only provide a short investigative opportunity into a process that continues for several months to years [90]. In vivo models of scarring and wound healing are ideal for investigative purposes as they provide the closest resemblance, if not the same,

216

Skin Tissue Models

to patients. However, limitations include difficulties in recruitment and current studies lacking consistency in study design, assessment techniques, and follow-up [82]. Tissue-engineered models are emerging to closely resemble the in vivo environment through regulation of environmental factors that would help resolve limitations with current wound-healing models [91,92]. Human cutaneous scar and wound models are critical in improving our understanding of the pathophysiology of cutaneous healing and scarring. They are additionally vital to aiding with the safe assessment of current and emerging wound therapeutics, with the ultimate aim of creating a scarring-free environment leading to optimal skin regeneration.

Abbreviations ECM ABM TGF-β1 LDH

extracellular matrix agent-based model transforming growth factor-beta 1 lactate dehydrogenase

References [1] Broughton G, Janis JE, Attinger CE. The basic science of wound healing. Plast Reconstr Surg 2006;117(7 Suppl):12S–34S. [2] Barrientos S, Stojadinovic O, Golinko MS, Brem H, Tomic-Canic M. Growth factors and cytokines in wound healing. Wound Repair Regen 2008;16(5):585–601. [3] van den Broek LJ, Limandjaja GC, Niessen FB, Gibbs S. Human hypertrophic and keloid scar models: principles, limitations and future challenges from a tissue engineering perspective. Exp Dermatol 2014;23(6):382–6. [4] Greenhalgh DG. Models of wound healing. J Burn Care Rehabil 2005;26(4):293–305. [5] Sidgwick GP, McGeorge D, Bayat A. Functional testing of topical skin formulations using an optimised ex vivo skin organ culture model. Arch Dermatol Res 2016;308(5):297–308. [6] Harunari N, Zhu KQ, Armendariz RT, Deubner H, Muangman P, Carrougher GJ, et al. Histology of the thick scar on the female, red duroc pig: final similarities to human ­hypertrophic scar. Burns 2006;32(6):669–77. [7] Simon GA, Maibach HI. The pig as an experimental animal model of percutaneous permeation in man: qualitative and quantitative observations—an overview. Skin Pharmacol Appl Skin Physiol 2000;13(5):229–34. [8] Pfuhler S, Fautz R, Ouedraogo G, Latil A, Kenny J, Moore C, et al. The cosmetics Europe strategy for animal-free genotoxicity testing: project status up-date. Toxicol In  Vitro 2014;28(1):18–23. [9] Ziraldo C, Mi Q, An G, Vodovotz Y. Computational modeling of inflammation and wound healing. Adv Wound Care (New Rochelle) 2013;2(9):527–37. [10] Waugh HV, Sherratt JA. Modeling the effects of treating diabetic wounds with engineered skin substitutes. Wound Repair Regen 2007;15(4):556–65. [11] Waugh  HV, Sherratt  JA. Macrophage dynamics in diabetic wound dealing. Bull Math Biol 2006;68(1):197–207. [12] Arciero JC, Mi Q, Branca MF, Hackam DJ, Swigon D. Continuum model of collective cell migration in wound healing and colony expansion. Biophys J 2011;100(3):535–43.

Wound healing and cutaneous scarring models of the human skin217

[13] Wearing HJ, Sherratt JA. Keratinocyte growth factor signalling: a mathematical model of dermal-epidermal interaction in epidermal wound healing. Math Biosci 2000;165(1):41–62. [14] McDougall S, Dallon J, Sherratt J, Maini P. Fibroblast migration and collagen deposition during dermal wound healing: mathematical modelling and clinical implications. Philos Trans A Math Phys Eng Sci 1843;2006(364):1385–405. [15] Mi Q, Rivière B, Clermont G, Steed DL, Vodovotz Y. Agent-based model of inflammation and wound healing: insights into diabetic foot ulcer pathology and the role of transforming growth factor-beta1. Wound Repair Regen 2007;15(5):671–82. [16] An G, Mi Q, Dutta-Moscato J, Vodovotz Y. Agent-based models in translational systems biology. Wiley Interdiscip Rev Syst Biol Med 2009;1(2):159–71. [17] Adra  S, Sun  T, MacNeil  S, Holcombe  M, Smallwood  R. Development of a three dimensional multiscale computational model of the human epidermis. PLoS One 2010;5(1):e8511. [18] Li X, Upadhyay AK, Bullock AJ, Dicolandrea T, Xu J, Binder RL, et al. Skin stem cell hypotheses and long term clone survival—explored using agent-based modelling. Sci Rep 2013;3:1904. [19] Sun T, McMinn P, Holcombe M, Smallwood R, MacNeil S. Agent based modelling helps in understanding the rules by which fibroblasts support keratinocyte colony formation. PLoS One 2008;3(5):e2129. [20] Sun  T, McMinn  P, Coakley  S, Holcombe  M, Smallwood  R, Macneil  S. An integrated systems biology approach to understanding the rules of keratinocyte colony formation. J R Soc Interface 2007;4(17):1077–92. [21] Anderson AR, Chaplain MA. Continuous and discrete mathematical models of tumor-­ induced angiogenesis. Bull Math Biol 1998;60(5):857–99. [22] McDougall  SR, Anderson  AR, Chaplain  MA, Sherratt  JA. Mathematical modelling of flow through vascular networks: implications for tumour-induced angiogenesis and chemotherapy strategies. Bull Math Biol 2002;64(4):673–702. [23] Chaplain MA, McDougall SR, Anderson AR. Mathematical modeling of tumor-induced angiogenesis. Annu Rev Biomed Eng 2006;8:233–57. [24] Gottrup F, Agren MS, Karlsmark T. Models for use in wound healing research: a survey focusing on in vitro and in vivo adult soft tissue. Wound Repair Regen 2000;8(2):83–96. [25] Cha D, O'Brien P, O'Toole EA, Woodley DT, Hudson LG. Enhanced modulation of keratinocyte motility by transforming growth factor-alpha (TGF-alpha) relative to epidermal growth factor (EGF). J Invest Dermatol 1996;106(4):590–7. [26] Bürk RR. A factor from a transformed cell line that affects cell migration. Proc Natl Acad Sci U S A 1973;70(2):369–72. [27] Buisson AC, Zahm JM, Polette M, Pierrot D, Bellon G, Puchelle E, et al. Gelatinase B is involved in the in vitro wound repair of human respiratory epithelium. J Cell Physiol 1996;166(2):413–26. [28] Calderon  M, Lawrence  WT, Banes  AJ. Increased proliferation in keloid fibroblasts wounded in vitro. J Surg Res 1996;61(2):343–7. [29] Henemyre-Harris CL, Adkins AL, Chuang AH, Graham JS. Addition of epidermal growth factor improves the rate of sulfur mustard wound healing in an in vitro model. Eplasty 2008;8:e16. [30] Kim WS, Lee JS, Bae GY, Kim JJ, Chin YW, Bahk YY, et al. Extract of Aneilema keisak inhibits transforming growth factor-β-dependent signalling by inducing Smad2 downregulation in keloid fibroblasts. Exp Dermatol 2013;22(1):69–71. [31] Butler PD, Ly DP, Longaker MT, Yang GP. Use of organotypic coculture to study keloid biology. Am J Surg 2008;195(2):144–8.

218

Skin Tissue Models

[32] Phan TT, Lim IJ, Aalami O, Lorget F, Khoo A, Tan EK, et al. Smad3 signalling plays an important role in keloid pathogenesis via epithelial-mesenchymal interactions. J Pathol 2005;207(2):232–42. [33] Chen WL, Kuo KT, Chou TY, Chen CL, Wang CH, Wei YH, et al. The role of cytochrome c oxidase subunit Va in non-small cell lung carcinoma cells: association with migration, invasion and prediction of distant metastasis. BMC Cancer 2012;12:273. [34] Harisi R, Kenessey I, Olah JN, Timar F, Babo I, Pogany G, et al. Differential inhibition of single and cluster type tumor cell migration. Anticancer Res 2009;29(8):2981–5. [35] Xie Y, Rizzi SC, Dawson R, Lynam E, Richards S, Leavesley DI, et al. Development of a three-dimensional human skin equivalent wound model for investigating novel wound healing therapies. Tissue Eng Part C Methods 2010;16(5):1111–23. [36] Eckes B, Krieg T, Nusgens BV, Lapière CM. In vitro reconstituted skin as a tool for biology, pharmacology and therapy: a review. Wound Repair Regen 1995;3(3):248–57. [37] Nayak S, Dey S, Kundu SC. Skin equivalent tissue-engineered construct: co-­cultured fibroblasts/keratinocytes on 3D matrices of sericin hope cocoons. PLoS One 2013;8(9):e74779. [38] Derderian  CA, Bastidas  N, Lerman  OZ, Bhatt  KA, Lin  SE, Voss  J, et  al. Mechanical strain alters gene expression in an in vitro model of hypertrophic scarring. Ann Plast Surg 2005;55(1):69–75 [discussion]. [39] Karamichos D, Lakshman N, Petroll WM. An experimental model for assessing fibroblast migration in 3-D collagen matrices. Cell Motil Cytoskeleton 2009;66(1):1–9. [40] Ghaffari  A, Kilani  RT, Ghahary  A. Keratinocyte-conditioned media regulate collagen expression in dermal fibroblasts. J Invest Dermatol 2009;129(2):340–7. [41] Bellemare J, Roberge CJ, Bergeron D, Lopez-Vallé CA, Roy M, Moulin VJ. Epidermis promotes dermal fibrosis: role in the pathogenesis of hypertrophic scars. J Pathol 2005;206(1):1–8. [42] Chiu LL, Sun CH, Yeh AT, Torkian B, Karamzadeh A, Tromberg B, et al. Photodynamic therapy on keloid fibroblasts in tissue-engineered keratinocyte-fibroblast co-culture. Lasers Surg Med 2005;37(3):231–44. [43] van den Broek LJ, Niessen FB, Scheper RJ, Gibbs S. Development, validation and testing of a human tissue engineered hypertrophic scar model. ALTEX 2012;29(4):389–402. [44] Bagabir R, Syed F, Paus R, Bayat A. Long-term organ culture of keloid disease tissue. Exp Dermatol 2012;21(5):376–81. [45] Duong  HS, Zhang  Q, Kobi  A, Le  A, Messadi  DV. Assessment of morphological and immunohistological alterations in long-term keloid skin explants. Cells Tissues Organs 2005;181(2):89–102. [46] Mendoza-Garcia  J, Sebastian  A, Alonso-Rasgado  T, Bayat  A. Ex  vivo evaluation of the effect of photodynamic therapy on skin scars and striae distensae. Photodermatol Photoimmunol Photomed 2015;31(5):239–51. [47] Syed F, Bagabir RA, Paus R, Bayat A. Ex vivo evaluation of antifibrotic compounds in skin scarring: EGCG and silencing of PAI-1 independently inhibit growth and induce keloid shrinkage. Lab Invest 2013;93(8):946–60. [48] Balaji S, Moles CM, Bhattacharya SS, LeSaint M, Dhamija Y, Le LD, et al. Comparison of interleukin 10 homologs on dermal wound healing using a novel human skin ex vivo organ culture model. J Surg Res 2014;190(1):358–66. [49] Companjen AR, van der Wel LI, Wei L, Laman JD, Prens EP. A modified ex vivo skin organ culture system for functional studies. Arch Dermatol Res 2001;293(4):184–90. [50] Peramo A, Marcelo CL, Goldstein SA, Martin DC. Improved preservation of the tissue surrounding percutaneous devices by hyaluronic acid and dermatan sulfate in a human skin explant model. Ann Biomed Eng 2010;38(3):1098–110.

Wound healing and cutaneous scarring models of the human skin219

[51] Reus AA, Usta M, Krul CA. The use of ex vivo human skin tissue for genotoxicity testing. Toxicol Appl Pharmacol 2012;261(2):154–63. [52] Tomic-Canic  M, Mamber  SW, Stojadinovic  O, Lee  B, Radoja  N, McMichael  J. Streptolysin O enhances keratinocyte migration and proliferation and promotes skin organ culture wound healing in vitro. Wound Repair Regen 2007;15(1):71–9. [53] Xu  W, Jong Hong  S, Jia  S, Zhao  Y, Galiano  RD, Mustoe  TA. Application of a ­partial-thickness human ex vivo skin culture model in cutaneous wound healing study. Lab Invest 2012;92(4):584–99. [54] Hodgkinson T, Bayat A. In vitro and ex vivo analysis of hyaluronan supplementation of Integra dermal template on human dermal fibroblasts and keratinocytes. J Appl Biomater Funct Mater 2016;14(1):e9–18. [55] Kratz  G. Modeling of wound healing processes in human skin using tissue culture. Microsc Res Tech 1998;42(5):345–50. [56] Moll I, Houdek P, Schmidt H, Moll R. Characterization of epidermal wound healing in a human skin organ culture model: acceleration by transplanted keratinocytes. J Invest Dermatol 1998;111(2):251–8. [57] Sebastian A, Iqbal SA, Colthurst J, Volk SW, Bayat A. Electrical stimulation enhances epidermal proliferation in human cutaneous wounds by modulating p53-SIVA1 interaction. J Invest Dermatol 2015;135(4):1166–74. [58] van Kilsdonk JW, van den Bogaard EH, Jansen PA, Bos C, Bergers M, Schalkwijk J. An in vitro wound healing model for evaluation of dermal substitutes. Wound Repair Regen 2013;21(6):890–6. [59] Ud-Din S, Bowring A, Derbyshire B, Morris J, Bayat A. Identification of steroid sensitive responders versus non-responders in the treatment of keloid disease. Arch Dermatol Res 2013;305(5):423–32. [60] Ud-Din S, Thomas G, Morris J, Bayat A. Photodynamic therapy: an innovative approach to the treatment of keloid disease evaluated using subjective and objective non-invasive tools. Arch Dermatol Res 2013;305(3):205–14. [61] Dunkin CS, Pleat JM, Gillespie PH, Tyler MP, Roberts AH, McGrouther DA. Scarring occurs at a critical depth of skin injury: precise measurement in a graduated dermal scratch in human volunteers. Plast Reconstr Surg 2007;119(6):1722–32 [discussion 33-4]. [62] Ud-Din S, Giddings Dip P, Colthurst J, Whiteside S, Morris J, Bayat A. Significant reduction of symptoms of scarring with electrical stimulation: evaluated with subjective and objective assessment tools in a prospective noncontrolled case series. Wounds 2013;25(8):212–24. [63] Ud-Din  S, McAnelly  SL, Bowring  A, Whiteside  S, Morris  J, Chaudhry  I, et  al. A ­double-blind controlled clinical trial assessing the effect of topical gels on striae distensae (stretch marks): a non-invasive imaging, morphological and immunohistochemical study. Arch Dermatol Res 2013;305(7):603–17. [64] Kiistala U. Dermal-epidermal separation. II. External factors in suction blister formation with special reference to the effect of temperature. Ann Clin Res 1972;4(4):236–46. [65] Lévy  JJ, von Rosen  J, Gassmüller  J, Kleine Kuhlmann  R, Lange  L. Validation of an in vivo wound healing model for the quantification of pharmacological effects on epidermal regeneration. Dermatology 1995;190(2):136–41. [66] Svedman P, Lundin S, Höglund P, Hammarlund C, Malmros C, Pantzar N. Passive drug diffusion via standardized skin mini-erosion; methodological aspects and clinical findings with new device. Pharm Res 1996;13(9):1354–9. [67] Jeong  SH, Kim  JH, Yi  SM, Lee  JP, Sohn  KH, Park  KL, et  al. Assessment of penetration of quantum dots through in  vitro and in  vivo human skin using the human skin

220

[68]

[69]

[70]

[71]

[72]

[73]

[74]

[75]

[76]

[77]

[78] [79]

[80] [81] [82]

[83]

Skin Tissue Models

equivalent model and the tape stripping method. Biochem Biophys Res Commun 2010;394(3):612–5. Brown  GL, Nanney  LB, Griffen  J, Cramer  AB, Yancey  JM, Curtsinger  LJ, et  al. Enhancement of wound healing by topical treatment with epidermal growth factor. N Engl J Med 1989;321(2):76–9. Greenhalgh  DG, Rieman  M. Effects of basic fibroblast growth factor on the healing of partial-thickness donor sites. A prospective, randomized, double-blind trial. Wound Repair Regen 1994;2(2):113–21. Bishop  T, Ballard  A, Holmes  H, Young  AR, McMahon  SB. Ultraviolet-B induced inflammation of human skin: characterisation and comparison with traditional models of hyperalgesia. Eur J Pain 2009;13(5):524–32. Mattsson U, Cassuto J, Tarnow P, Jönsson A, Jontell M. Intravenous lidocaine infusion in the treatment of experimental human skin burns—digital colour image analysis of erythema development. Burns 2000;26(8):710–5. Condé-Green A, Chung TL, Holton LH, Hui-Chou HG, Zhu Y, Wang H, et al. Incisional negative-pressure wound therapy versus conventional dressings following abdominal wall reconstruction: a comparative study. Ann Plast Surg 2013;71(4):394–7. Ud-Din S, Sebastian A, Giddings P, Colthurst J, Whiteside S, Morris J, et al. Angiogenesis is induced and wound size is reduced by electrical stimulation in an acute wound healing model in human skin. PLoS One 2015;10(4):e0124502. Ud-Din S, Greaves NS, Sebastian A, Baguneid M, Bayat A. Noninvasive device readouts validated by immunohistochemical analysis enable objective quantitative assessment of acute wound healing in human skin. Wound Repair Regen 2015;23(6):901–14. Greaves NS, Lqbal SA, Morris J, Benatar B, Alonso-Rasgado T, Baguneid M, et al. Acute cutaneous wounds treated with human decellularised dermis show enhanced angiogenesis during healing. PLoS One 2015;10(1):e0113209. Greaves NS, Iqbal SA, Hodgkinson T, Morris J, Benatar B, Alonso-Rasgado T, et al. Skin substitute-assisted repair shows reduced dermal fibrosis in acute human wounds validated simultaneously by histology and optical coherence tomography. Wound Repair Regen 2015;23(4):483–94. Ud-Din S, Perry D, Giddings P, Colthurst J, Zaman K, Cotton S, et al. Electrical stimulation increases blood flow and haemoglobin levels in acute cutaneous wounds without affecting wound closure time: evidenced by non-invasive assessment of temporal biopsy wounds in human volunteers. Exp Dermatol 2012;21(10):758–64. Reid RR, Said HK, Mogford JE, Mustoe TA. The future of wound healing: pursuing surgical models in transgenic and knockout mice. J Am Coll Surg 2004;199(4):578–85. Falanga  V, Schrayer  D, Cha  J, Butmarc  J, Carson  P, Roberts  AB, et  al. Full-thickness wounding of the mouse tail as a model for delayed wound healing: accelerated wound closure in Smad3 knock-out mice. Wound Repair Regen 2004;12(3):320–6. Grose  R, Werner  S. Wound-healing studies in transgenic and knockout mice. Mol Biotechnol 2004;28(2):147–66. Dunn L, Prosser HC, Tan JT, Vanags LZ, Ng MK, Bursill CA. Murine model of wound healing. J Vis Exp 2013;75:e50265. Ud-Din S, Bayat A. Non-animal models of wound healing in cutaneous repair: in silico, in vitro, ex vivo and in vivo models of wounds and scars in human skin. Wound Rep Reg 2017;25:164–76. https://doi.org/10.1111/wrr.12513. Lindblad  WJ. Considerations for selecting the correct animal model for dermal wound-healing studies. J Biomater Sci Polym Ed 2008;19(8):1087–96.

Wound healing and cutaneous scarring models of the human skin221

[84] Roy S, Biswas S, Khanna S, Gordillo G, Bergdall V, Green J, et al. Characterization of a preclinical model of chronic ischemic wound. Physiol Genomics 2009;37(3):211–24. [85] Singer AJ, McClain SA, Taira BR, Romanov A, Rooney J, Zimmerman T. Validation of a porcine comb burn model. Am J Emerg Med 2009;27(3):285–8. [86] Navarro FA, Stoner ML, Park CS, Huertas JC, Lee HB, Wood FM, et al. Sprayed keratinocyte suspensions accelerate epidermal coverage in a porcine microwound model. J Burn Care Rehabil 2000;21(6):513–8. [87] Elgharably H, Roy S, Khanna S, Abas M, Dasghatak P, Das A, et al. A modified collagen gel enhances healing outcome in a preclinical swine model of excisional wounds. Wound Repair Regen 2013;21(3):473–81. [88] Montagna W, Yun JS. The skin of the domestic pig. J Invest Dermatol 1964;42:11–21. [89] Vermolen FJ, Javierre E. Computer simulations from a finite-element model for wound contraction and closure. J Tissue Viability 2010;19(2):43–53. [90] Hewitt  NJ, Edwards  RJ, Fritsche  E, Goebel  C, Aeby  P, Scheel  J, et  al. Use of human in vitro skin models for accurate and ethical risk assessment: metabolic considerations. Toxicol Sci 2013;133(2):209–17. [91] Ataç B, Wagner I, Horland R, Lauster R, Marx U, Tonevitsky AG, et al. Skin and hair ona-chip: in vitro skin models versus ex vivo tissue maintenance with dynamic perfusion. Lab Chip 2013;13(18):3555–61. [92] Wagner I, Materne EM, Brincker S, Süssbier U, Frädrich C, Busek M, et al. A dynamic multi-organ-chip for long-term cultivation and substance testing proven by 3D human liver and skin tissue co-culture. Lab Chip 2013;13(18):3538–47.